The S&P 500 Index Prediction Based on N-BEATS
Yichen Liu, Chengcheng Zhong, Qiaoyu Ma, Yanan Jiang, Chunlei Zhang · Atlantis Highlights in Computer Sciences/Atlantis highlights in computer sciences · 2023
The stock market prediction has been a hot topic in the field of economics and finance.As a consequence of the complex and volatile nature of the stock market, it is challenging to accurately forecast the stock S&P 500 index.Currently, with the purpose of predicting stock market, intelligent algorithms via computer have been proved superior in recent studies.We have introduced the N-BEATS algorithm to precisely estimate the stock S&P 500 index which are tailored towards the drawbacks that most algorithms cannot incorporate with historical information for time-series data.The features extracted by the N-BEATS algorithm are more consistent with the temporal features through the forward and backward coefficients.On the basis of the comparison of four evaluation metrics obtained from the S&P 500 index corresponding to 500 base stocks in this study, the N-BEATS algorithm outperforms other estimators.It can be demonstrated that the N-BEATS is a more suitable and promising method for stock market prediction, which has widespread application value.